The literature on team cooperation has neglected the effects of relative kindness intention on cooperation, which we measure by comparing the kindness intentions of an agent to her group members to the kindness shown by other members to this same agent. We argue that the agent's emotional reaction to material payoff inequity is not constant, but rather affected by her relative kindness intention. Then, we apply the model to team projects with multiple partners and investigate how inequity-aversion and relative kindness intention jointly influence team cooperation. We first consider the case of homogeneous agents, where their marginal productivity levels and technical capacities are the same, and then consider the case of heterogeneous agents, where their marginal productivity levels and technical capacities are not the same. Our results show that inequity-aversion has no effect on effort expenditure in the former case, but does affect it in the latter case. The consideration of relative kindness intention may impact the agents' optimal cooperative effort expenditure when their technical capacities are different. In addition, it is beneficial for team cooperation, and might not only reduce the negative impact but also enhance the positive impact of inequity-aversion on the agents' effort expenditures.

Compassion, the emotional response of caring for another who is suffering and that results in motivation to relieve suffering, is thought to be an emotional antecedent to altruistic behavior. However, it remains unclear whether compassion enhances altruistic behavior in a uniform way or is specific to sub-types of behavior such as altruistic helping of a victim or altruistic punishment of a transgressor. We investigated the relationship between compassion and subtypes of altruistic behavior using third-party paradigms where participants (1) witnessed an unfair economic exchange between a transgressor and a victim, and (2) had the opportunity to either spend personal funds to either economically (a) help the victim or (b) punish the transgressor. In Study 1, we examined whether individual differences in self-reported empathic concern (the emotional component of compassion) was associated with greater altruistic helping or punishment behavior in two independent samples. For participants who witnessed an unfair transaction, trait empathic concern was associated with greater helping of a victim and had no relationship to punishment. However, in those who decided to punish the transgressor, participants who reported greater empathic concern decided to punish less. In Study 2, we directly enhanced compassion using short-term online compassion meditation training to examine whether altruistic helping and punishment were increased after two weeks of training. Compared to an active reappraisal training control group, the compassion training group gave more to help the victim and did not differ in punishment of the transgressor. Together, these two studies suggest that compassion is related to greater altruistic helping of victims and is not associated with or may mitigate altruistic punishment of transgressors.